Method and device for predicting unsteady productivity of closed boundary fracturing horizontal well

Abstract

The present application relates to the technical field of data analysis and prediction, and particularly relates to a method and device for predicting unsteady-state productivity of a closed boundary fractured horizontal well. The method comprises the following steps: collecting data of the closed boundary fractured horizontal well to generate fractured horizontal well data; pre-processing the fractured horizontal well data to obtain standard fractured horizontal well data; identifying the closed boundary fracturing characteristics of the standard fractured horizontal well data to obtain closed boundary fracturing data; extracting the fracture characteristics of the closed boundary fracturing data to generate boundary fracture characteristic information; the present application realizes the identification of the fracture mutual relationship between the bottom hole pressure and the boundary fracture characteristics through data processing technology, pattern recognition technology and deep learning technology; and realizes the detection of the oil and gas flow state according to the oil and gas characteristics based on the boundary fracture relationship, thereby improving the accuracy of unsteady-state productivity prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data analysis and prediction, and particularly relates to a method and device for predicting non-steady-state productivity of a closed boundary fracturing horizontal well. BACKGROUND

[0002] For the prediction of non-steady-state productivity of a horizontal well, initially, fracturing stimulation technology is used, specifically including hydraulic jetting staged fracturing technology, open hole packer staged fracturing technology and fast drilling bridge plug staged fracturing technology, and the open hole packer staged fracturing technology is widely used in the measurement and analysis of non-steady-state productivity of oil and gas; with the improvement of fracturing technology, especially the closed boundary fracturing technology, the wellsite coordinates of all mirror wells corresponding to the closed boundary formation fracturing equivalent well of the equivalent well can be determined according to the wellsite coordinates of the equivalent well, and then the distances from all mirror wells to the equivalent well can be determined; so that the non-steady-state productivity can be predicted according to relevant parameters; however, the existing technology fails to identify the fracture relationship between the bottom hole pressure and the boundary fracture characteristics; and fails to detect the oil and gas flow state according to the boundary fracture relationship, thereby resulting in low accuracy of non-steady-state productivity prediction. SUMMARY

[0003] Therefore, it is necessary to provide a method and device for predicting non-steady-state productivity of a closed boundary fracturing horizontal well, so as to solve at least one of the above technical problems.

[0004] To achieve the above-mentioned purpose, a method for predicting non-steady-state productivity of a closed boundary fracturing horizontal well, the method comprising the following steps:

[0005] Step S1: collecting data of the closed boundary fracturing horizontal well to generate fracturing horizontal well data; and pre-processing the fracturing horizontal well data to obtain standard fracturing horizontal well data;

[0006] Step S2: identifying closed boundary fracturing characteristics of the standard fracturing horizontal well data to obtain closed boundary fracturing data; extracting fracture characteristics of the closed boundary fracturing data to generate boundary fracture characteristic information; determining fracturing porosity of the standard fracturing horizontal well data according to the boundary fracture characteristic information to obtain fracturing porosity data; and simulating oil and gas characteristics of the standard fracturing horizontal well data based on the boundary fracture characteristic information and the fracturing porosity data to generate oil and gas characteristic data;

[0007] Step S3: crack parameter extraction is performed on the boundary crack feature information through the oil and gas characteristic data to obtain boundary crack parameter information; wellbore position data is obtained by identifying the wellbore position of the standard fracturing horizontal well data according to the boundary crack parameter information; crack wellbore correlation data is generated by correlating and mapping the boundary crack parameter information and the wellbore position data; and the bottom hole pressure data is obtained by determining the bottom hole pressure of the standard fracturing horizontal well data based on the crack wellbore correlation data.

[0008] Step S4: boundary crack mutual coupling analysis is performed on the boundary crack feature information according to the bottom hole pressure data to generate boundary crack relationship data; oil and gas flow state data is obtained by identifying the oil and gas flow state of the oil and gas characteristic data according to the boundary crack relationship data; oil and gas unsteady-state productivity index is generated by measuring and calculating the oil and gas unsteady-state productivity index of the standard fracturing horizontal well data through the oil and gas flow state data; the unsteady-state productivity prediction model is constructed by using the oil and gas unsteady-state productivity index, and the unsteady-state productivity prediction model is obtained; and the unsteady-state productivity prediction report is generated by predicting the unsteady-state productivity of the standard fracturing horizontal well data based on the unsteady-state productivity prediction model.

[0009] The application uses a sensor to collect data of a closed boundary fracturing horizontal well, ensures to obtain original and direct information about the fracturing horizontal well, can provide basic information for subsequent analysis, pre-processes the fracturing horizontal well data, helps to eliminate noise and abnormal values in the data, can ensure consistency and accuracy of the data, and provides key data input for subsequent fracturing porosity determination and oil and gas feature simulation. The closed boundary fracturing feature recognition of the standard fracturing horizontal well data can evaluate key of the fracture network and reservoir connectivity, the fracture feature extraction of the closed boundary fracturing data can clearly determine the boundary fracture feature information, the fracturing porosity determination of the standard fracturing horizontal well data according to the boundary fracture feature information can accurately reflect the pore structure of the fracturing region, which is crucial for evaluating the reservoir characteristics of the oil and gas reservoir, the oil and gas feature simulation of the standard fracturing horizontal well data based on the boundary fracture feature information and the fracturing porosity data can simulate the distribution and flow of oil and gas in the fracturing horizontal well. The fracture parameter extraction of the boundary fracture feature information through the oil and gas feature data can describe the geometric characteristics and distribution characteristics of the fracture in detail, the wellbore position recognition of the standard fracturing horizontal well data according to the boundary fracture parameter information can determine the accurate position of the wellbore in the oil and gas reservoir, the correlation mapping of the boundary fracture parameter information and the wellbore position data can correlate the fracture parameters and the wellbore position, and provides data support for the bottom hole pressure determination, the bottom hole pressure determination of the standard fracturing horizontal well data based on the fracture wellbore correlation data can clearly determine the bottom hole pressure data. The boundary fracture mutual coupling analysis of the boundary fracture feature information according to the bottom hole pressure data can reveal the interaction and influence between the fractures, the oil and gas flow state recognition of the oil and gas feature data according to the boundary fracture relationship data can identify the flow state of oil and gas in the fracture network, the oil and gas non-steady state productivity index measurement of the standard fracturing horizontal well data through the oil and gas flow state data can provide dynamic change information of the oil and gas production, the non-steady state productivity prediction model construction using the oil and gas non-steady state productivity index can obtain a non-steady state productivity prediction model, and the non-steady state productivity prediction of the standard fracturing horizontal well data based on the non-steady state productivity prediction model can generate a non-steady state productivity prediction report. Therefore, through the data processing technology, the pattern recognition technology, and the deep learning technology, the application realizes the fracture mutual relationship recognition of the bottom hole pressure and the boundary fracture feature, and realizes the oil and gas flow state detection of the oil and gas feature according to the boundary fracture relationship, so as to improve the accuracy of the non-steady state productivity prediction.

[0010] Preferably, the step S2 comprises the following steps:

[0011] Step S21: performing nonlinear feature decomposition on the standard fracturing horizontal well data to generate nonlinear fracturing horizontal well features; performing fracturing feature enhancement on the nonlinear fracturing horizontal well features to obtain fracturing feature enhancement data; performing closed boundary fracturing feature extraction on the fracturing feature enhancement data to obtain closed boundary fracturing data;

[0012] Step S22: performing fracturing mode feature extraction on the closed boundary fracturing data to generate fracturing mode feature data; performing geometric feature quantification on the standard fracturing horizontal well data to obtain fracturing mode geometric features; performing fracture feature identification on the fracturing mode geometric features to generate boundary fracture feature information;

[0013] Step S23: performing distribution range determination on the boundary fracture feature information to obtain a boundary fracture distribution range; performing fracture spatial distribution feature identification on the boundary fracture distribution range to generate a fracture spatial distribution feature; performing fracture spatial distribution mode feature on the fracturing mode geometric features according to the fracture spatial distribution feature to obtain a fracture spatial distribution mode feature;

[0014] Step S24: performing fracture density calculation on the fracture spatial distribution mode feature to generate fracture spatial density data; performing fracturing pore feature on the boundary fracture feature information based on the fracture spatial density data to obtain fracturing pore features; performing fracturing porosity measurement on the standard fracturing horizontal well data through the fracturing pore features to obtain fracturing porosity data;

[0015] Step S25: performing porosity gradient division on the fracturing porosity data to generate porosity gradient data; performing oil and gas feature simulation on the standard fracturing horizontal well data based on the porosity gradient data and the boundary fracture feature information to generate oil and gas feature data.

[0016] The application can reveal the complex mode and trend in the data by performing nonlinear feature decomposition on the standard fracturing horizontal well data; can improve the recognition degree of key features in the data by performing fracturing feature enhancement on the nonlinear fracturing horizontal well features; can clearly identify the fracturing features related to the closed boundary by performing closed boundary fracturing feature extraction on the fracturing feature enhanced data; can describe the geometric shape and distribution of the fracturing cracks in detail by performing fracturing morphology feature extraction on the closed boundary fracturing data; can accurately describe the properties of the cracks by performing crack feature identification; can clearly determine the extension range of the cracks in the oil and gas reservoir by determining the distribution range of the boundary crack feature information; can reveal the spatial distribution law of the cracks by performing crack spatial distribution feature identification on the boundary crack distribution range; can refine the spatial distribution features of the cracks by performing crack spatial distribution morphology feature on the fracturing morphology geometric features according to the crack spatial distribution features; can make the quantitative crack density calculation by performing crack density calculation on the crack spatial distribution morphology features; can describe the influence of the cracks on the pore structure by performing fracturing pore feature on the boundary crack feature information based on the crack spatial density data; can clearly determine the fracturing porosity data by performing fracturing porosity measurement on the standard fracturing horizontal well data through the fracturing pore feature; can reveal the spatial variation of the porosity in the oil and gas by performing porosity gradient division on the fracturing porosity data; and can simulate the distribution and flow of the oil and gas in the crack network by performing oil and gas feature simulation on the standard fracturing horizontal well data based on the porosity gradient data and the boundary crack feature information.

[0017] Preferably, step S3 comprises the following steps:

[0018] Step S31: performing crack geometric feature identification on the boundary crack feature information to generate crack geometric feature data; and performing crack geometric parameter extraction on the crack geometric feature data according to the oil and gas feature data to obtain crack geometric parameters;

[0019] Step S32: performing crack connectivity detection on the boundary crack feature information through the crack geometric parameters to generate crack connectivity data; performing crack permeability evaluation on the boundary crack feature information based on the crack connectivity data to obtain crack permeability data; and performing crack fluid flow simulation on the oil and gas feature data according to the crack permeability data to obtain crack fluid flow simulation data;

[0020] Step S33: integrating the crack connectivity data, the crack permeability data and the crack fluid flow simulation data to obtain boundary crack parameter information; performing wellbore feature identification on the standard fracturing horizontal well data to generate wellbore feature data; and performing wellbore position detection on the wellbore feature data according to the boundary crack parameter information to obtain wellbore position data;

[0021] Step S34: mapping the boundary fracture parameter information and the wellbore position data to generate fracture wellbore correlation data; and determining the bottom hole pressure based on the fracture wellbore correlation data.

[0022] The present application can describe the geometry of the fracture in detail, including the strike, dip and dip angle of the fracture, by identifying the fracture geometry characteristics of the boundary fracture feature information; can quantify the size and shape of the fracture by extracting the fracture geometry parameters from the fracture geometry characteristics data according to the oil and gas feature data; can reveal the connection relationship between the fractures by detecting the fracture connectivity of the boundary fracture feature information based on the fracture connectivity data; can quantitatively determine the permeability of the fracture by evaluating the fracture permeability of the boundary fracture feature information based on the fracture connectivity data; can clearly simulate the flow behavior of oil and gas in the fracture by simulating the fracture fluid flow of the oil and gas feature data according to the fracture permeability data; can comprehensively describe the characteristics of the fracture by integrating the fracture connectivity data, the fracture permeability data and the fracture fluid flow simulation data, thereby providing key data support for the wellbore feature identification and the wellbore position detection; can determine the geometry and spatial position of the wellbore by identifying the wellbore feature of the standard fracturing horizontal well data; can determine the specific position of the wellbore in the oil and gas reservoir by detecting the wellbore position of the wellbore feature data according to the boundary fracture parameter information; can generate the fracture wellbore correlation data by mapping the boundary fracture parameter information and the wellbore position data; and can directly reflect the dynamic pressure condition of the bottom hole by determining the bottom hole pressure based on the fracture wellbore correlation data.

[0023] Preferably, step S34 comprises the following steps:

[0024] Step S341: quantifying the fracture connectivity data to obtain fracture connectivity quantitative data; and correlating the fracture connectivity quantitative data and the wellbore position data to generate fracture wellbore connectivity correlation data;

[0025] Step S342: parameterizing the fracture permeability data to obtain fracture permeability parameterized data; and correlating the fracture permeability parameterized data and the wellbore position data to generate fracture wellbore permeability correlation data;

[0026] Step S343: extracting the flow characteristics of the fracture fluid flow simulation data to obtain fracture fluid flow characteristic data; and correlating the fracture fluid flow characteristic data and the wellbore position data to generate fracture wellbore flowability correlation data;

[0027] Step S344: integrating the fracture wellbore connectivity correlation data, the fracture wellbore permeability correlation data and the fracture wellbore flowability correlation data to obtain fracture wellbore correlation data;

[0028] Step S345: based on the fracture wellbore correlation data, the bottom hole pressure of the standard fractured horizontal well data is determined to obtain the bottom hole pressure data.

[0029] The present application quantifies the fracture connectivity data, can provide an accurate numerical value for the connectivity of the fracture network; the fracture connectivity quantitative data is correlated with the wellbore position data to determine the connectivity between the fracture and the wellbore; the fracture permeability data is parameterized to determine the permeability characteristics of the fracture; the fracture permeability parameterized data is correlated with the wellbore position data to reveal the influence of the fracture permeability on the wellbore flow characteristics; the fracture fluid flow simulation data is extracted to simulate the flow behavior of oil and gas in the fracture; the fracture fluid flow characteristic data is correlated with the wellbore position data to combine the flow characteristics in the fracture with the wellbore position information; the fracture wellbore connectivity correlation data, the fracture wellbore permeability correlation data and the fracture wellbore flow correlation data are integrated to comprehensively describe the interaction between the fracture and the wellbore; based on the fracture wellbore correlation data, the bottom hole pressure of the standard fractured horizontal well data is determined to directly reflect the dynamic pressure condition of the bottom hole.

[0030] Preferably, step S345 includes the following steps:

[0031] Step S3451: based on the fracture wellbore correlation data, the bottom hole stress characteristics of the standard fractured horizontal well data are extracted to generate the bottom hole stress characteristics; the bottom hole pressure distribution of the standard fractured horizontal well data is simulated according to the bottom hole stress characteristics to obtain the bottom hole pressure distribution simulation data;

[0032] Step S3452: the bottom hole pressure response data is generated by identifying the bottom hole pressure response of the bottom hole pressure distribution simulation data; the bottom hole pressure numerical information is obtained by calculating the pressure value of the bottom hole pressure response data; the bottom hole pressure gradient data is generated by dividing the pressure gradient of the bottom hole pressure numerical information;

[0033] Step S3453: the real-time bottom hole pressure of the bottom hole pressure distribution simulation data is monitored through the bottom hole pressure gradient data to obtain the bottom hole pressure change data; the bottom hole pressure state information is generated by identifying the bottom hole pressure state of the bottom hole pressure change data;

[0034] Step S3454: the bottom hole pressure anomaly data is obtained by detecting the bottom hole pressure anomaly of the bottom hole pressure state information; the bottom hole pressure stability data is obtained by evaluating the pressure stability of the bottom hole pressure distribution simulation data according to the bottom hole pressure anomaly data; the bottom hole pressure data is obtained by determining the bottom hole pressure of the standard fractured horizontal well data based on the bottom hole pressure stability data.

[0035] The application can clearly determine the action of various forces on the well bottom according to the well bottom stress feature extraction of the standard fracturing horizontal well data based on the fracture wellbore correlation data; can simulate the pressure state of different regions of the well bottom according to the well bottom pressure distribution simulation of the standard fracturing horizontal well data based on the well bottom stress feature; can reflect the sensitivity of the well bottom to the pressure change in the oil and gas production process according to the well bottom pressure response identification of the well bottom pressure distribution simulation data; can provide the specific value of the well bottom pressure according to the pressure numerical calculation of the well bottom pressure response data; can reveal the change trend of the well bottom pressure in different regions according to the pressure gradient division of the well bottom pressure numerical information; can reflect the dynamic change of the well bottom pressure in real time according to the real-time well bottom pressure monitoring of the well bottom pressure distribution simulation data based on the well bottom pressure gradient data; can clearly determine the current state of the well bottom pressure according to the well bottom pressure state identification of the well bottom pressure change data; can timely find the abnormal change of the well bottom pressure according to the well bottom pressure anomaly detection of the well bottom pressure state information; can evaluate the stability of the well bottom pressure according to the pressure stability evaluation of the well bottom pressure distribution simulation data based on the well bottom pressure anomaly data; and can determine the well bottom pressure data based on the well bottom pressure measurement of the standard fracturing horizontal well data based on the well bottom pressure stability data.

[0036] Preferably, step S4 comprises the following steps:

[0037] Step S41: boundary fracture mutual coupling analysis is performed on the boundary fracture feature information according to the well bottom pressure data, and boundary fracture relationship data is generated;

[0038] Step S42: oil and gas flow state identification is performed on the oil and gas feature data according to the boundary fracture relationship data, and oil and gas flow state data is obtained;

[0039] Step S43: oil and gas non-steady-state productivity index measurement is performed on the standard fracturing horizontal well data through the oil and gas flow state data, and oil and gas non-steady-state productivity index is generated;

[0040] Step S44: a non-steady-state productivity prediction model is constructed according to the oil and gas non-steady-state productivity index, and a non-steady-state productivity prediction training model is obtained by training the non-steady-state productivity prediction model using the oil and gas non-steady-state productivity index;

[0041] Step S45: model cross-validation evaluation is performed on the non-steady-state productivity prediction training model, and model evaluation data is obtained; model parameter adjustment is performed on the non-steady-state productivity prediction training model through the model evaluation data, and a non-steady-state productivity prediction model is obtained;

[0042] Step S46: non-steady-state productivity prediction is performed on the standard fracturing horizontal well data based on the non-steady-state productivity prediction model, and a non-steady-state productivity prediction report is generated.

[0043] The application can clearly reflect the mutual relationship and interaction between the cracks according to the boundary crack mutual coupling analysis of the boundary crack characteristic information according to the well bottom pressure data; the flow characteristics of the oil and gas in the crack network can be accurately reflected, specifically including the flow direction, speed and flow mode of the oil and gas, according to the oil and gas flow state recognition of the oil and gas characteristic data according to the boundary crack relationship data; the production capacity of the oil and gas reservoir in different production stages can be reflected, which can quantify the production capacity change of the oil and gas reservoir, according to the oil and gas unsteady state productivity index measurement of the standard fracturing horizontal well data through the oil and gas flow state data; the unsteady state productivity prediction model can be obtained, which can predict the production capacity change of the oil and gas at different time points in the future, according to the unsteady state productivity prediction model construction of the oil and gas unsteady state productivity index; the prediction accuracy can be improved by training the unsteady state productivity prediction model using the oil and gas unsteady state productivity index; the prediction performance and generalization ability of the model can be evaluated according to the model cross-validation evaluation of the unsteady state productivity prediction training model; the actual production capacity change of the oil and gas reservoir can be more accurately reflected, and the reliability of the prediction result can be improved, according to the model parameter adjustment of the unsteady state productivity prediction training model through the model evaluation data; the unsteady state productivity prediction of the standard fracturing horizontal well data can be performed based on the unsteady state productivity prediction model, and the unsteady state productivity prediction report can be generated, which can perform the unsteady state productivity prediction operation of the closed boundary fracturing horizontal well.

[0044] Preferably, step S41 comprises the following steps:

[0045] Step S411: extracting pressure distribution characteristics from the well bottom pressure data to obtain well bottom pressure distribution characteristics; identifying the pressure peak area of the well bottom pressure distribution characteristics to generate the well bottom pressure peak area;

[0046] Step S412: identifying the crack concentration area of the boundary crack characteristic information according to the crack spatial density data to obtain the boundary crack concentration area; mapping and matching the boundary crack concentration area and the well bottom pressure peak area to generate crack pressure mapping data;

[0047] Step S413: calculating the crack pressure influence degree of the well bottom pressure data through the crack pressure mapping data to obtain crack pressure influence data; determining the crack sensitivity of the boundary crack characteristic information according to the crack pressure influence data to generate crack sensitivity data;

[0048] Step S414: extracting the boundary crack constraint characteristic of the crack sensitivity data using the crack connectivity data to obtain the boundary crack constraint action data; extracting the boundary crack permeability synergy characteristic of the boundary crack constraint action data through the crack permeability data to generate the boundary crack synergy action data; identifying the boundary crack mutual coupling of the boundary crack characteristic information according to the boundary crack constraint action data and the boundary crack synergy action data to obtain the boundary crack relationship data.

[0049] The well bottom pressure data is subjected to pressure distribution feature extraction, the distribution of the well bottom pressure can be determined; the well bottom pressure distribution features are subjected to pressure peak region identification, the well bottom high pressure region can be accurately located; the boundary fracture feature information is subjected to fracture concentration region identification according to the fracture spatial density data, the main region of fracture development can be determined; the boundary fracture concentration region and the well bottom pressure peak region are subjected to mapping matching, which can be used for subsequent evaluation of the influence of the fracture on the well bottom pressure; the fracture pressure influence degree of the well bottom pressure data is calculated through the fracture pressure mapping data, the contribution of the fracture to the well bottom pressure can be quantified; the fracture sensitivity is determined according to the fracture pressure influence data of the boundary fracture feature information, the sensitivity of the fracture to the pressure change can be revealed; the boundary fracture constraint feature extraction is performed on the fracture sensitivity data by using the fracture connectivity data, the interaction and constraint relationship between the fractures can be described; the boundary fracture permeability synergy feature extraction is performed on the boundary fracture constraint action data by using the fracture permeability data, the synergy effect of the fracture permeability on the oil and gas flow can be revealed; the boundary fracture mutual coupling is identified according to the boundary fracture constraint action data and the boundary fracture synergy action data, the mutual relationship between the fractures can be comprehensively described.

[0050] Preferably, step S42 comprises the following steps:

[0051] Step S421: according to the boundary fracture relationship data, the fracture-oil and gas flow action data is obtained by corresponding the oil and gas feature data to the fracture-oil and gas flow action;

[0052] Step S422: the oil and gas flow feature data is labeled to generate oil and gas flow label data; the flow unit division data is obtained by dividing the oil and gas feature data into flow units through the oil and gas flow label data; the oil and gas flow characteristic data is generated by extracting the flow characteristics of the flow unit division data;

[0053] Step S423: the fracture-oil and gas flow network is constructed based on the fracture-oil and gas flow action data, the oil and gas flow direction path data is generated by identifying the oil and gas flow direction path of the fracture-oil and gas flow network; the oil and gas flow direction flow data is obtained by calculating the oil and gas flow direction flow of the fracture-oil and gas flow network according to the oil and gas flow direction path data;

[0054] Step S424: the flow state abnormality data is generated by detecting the flow state abnormality of the fracture-oil and gas flow network through the oil and gas flow direction flow data; the flow state stability data is obtained by evaluating the flow state stability of the fracture-oil and gas flow network;

[0055] Step S425: integrating the oil and gas flow direction flow data, the oil and gas flow state anomaly data and the flow state stability data to obtain oil and gas flow state data.

[0056] According to the boundary crack relationship data, the oil and gas characteristic data are subjected to crack-oil and gas flow action correspondence, the interaction relationship between the cracks and the oil and gas flow can be determined; the oil and gas characteristic data are subjected to oil and gas flow characteristic marking, different modes and stages of the oil and gas flow can be identified and distinguished; the flow unit division is performed on the oil and gas characteristic data through the oil and gas flow marking data, specifically, the oil and gas reservoir is divided into regions with similar flow characteristics; the flow characteristic extraction is performed on the flow unit division data, the flow characteristics of each flow unit are determined, including the flow speed, direction and flow efficiency; the crack-oil and gas flow network is constructed based on the crack-oil and gas flow action data, the flow path and flow efficiency of the oil and gas in the cracks can be simulated; the oil and gas flow direction path is identified through the crack-oil and gas flow network, the specific path of the oil and gas flow can be determined; the oil and gas flow direction flow is calculated through the crack-oil and gas flow network based on the oil and gas flow direction path data, the scale of the oil and gas flow can be quantified; the flow state anomaly detection is performed on the crack-oil and gas flow network through the oil and gas flow direction flow data, the abnormal situation in the oil and gas flow can be found in time; the flow state stability evaluation is performed on the crack-oil and gas flow network, the stability of the oil and gas flow can be evaluated; the oil and gas flow state data are obtained by integrating the oil and gas flow direction flow data, the oil and gas flow state anomaly data and the flow state stability data.

[0057] Preferably, step S43 comprises the following steps:

[0058] Step S431: unit oil and gas volume statistics are performed on the oil and gas flow direction flow data to obtain unit oil and gas volume data; the instantaneous oil and gas production is calculated based on the unit oil and gas volume data to generate instantaneous oil and gas production data;

[0059] Step S432: the flow anomaly state type is obtained by performing anomaly state type identification on the oil and gas flow state anomaly data; the abnormal instantaneous production data are generated by performing abnormal instantaneous production recording on the instantaneous oil and gas production data based on the flow anomaly state type; the abnormal state cumulative production data are obtained by performing oil and gas production accumulation on the abnormal state instantaneous production data;

[0060] Step S433: the stable state instantaneous production data are generated by performing stable state instantaneous production recording on the instantaneous oil and gas production data based on the flow state stability data; the stable state cumulative production data are obtained by performing oil and gas production accumulation on the stable state instantaneous production data;

[0061] Step S434: the oil and gas non-steady state productivity index is obtained by performing production value difference on the stable state cumulative production data and the abnormal state cumulative production data.

[0062] The application can quantify the flow scale of oil and gas by unit oil and gas volume statistics of oil and gas flow rate data; can clarify the instantaneous oil and gas production information at a specific time point by instantaneous oil and gas production calculation of unit oil and gas volume data; can clearly distinguish the abnormal state in the oil and gas flow by abnormal state type identification of oil and gas flow state abnormal data, which is crucial for timely identifying and responding to potential problems of oil and gas wells; can record the oil and gas production under abnormal conditions by abnormal instantaneous production recording of instantaneous oil and gas production data according to the flow abnormal state type; can clarify the total production in a period of time due to abnormal state by oil and gas production accumulation of abnormal state instantaneous production data; can accurately reflect the production of oil and gas wells under normal state by stable state instantaneous production recording of instantaneous oil and gas production data based on flow state stability data; can clarify the total normal production of oil and gas wells in a period of time by oil and gas production accumulation of stable state instantaneous production data; and can quantify the productivity change of oil and gas wells under non-steady state conditions by making difference between the stable state cumulative production data and the abnormal state cumulative production data.

[0063] In the present specification, a closed boundary fracturing horizontal well non-steady state productivity prediction device is provided, which comprises a prediction system for performing the closed boundary fracturing horizontal well non-steady state productivity prediction method described above, and the prediction system comprises:

[0064] A data acquisition module is configured to acquire data of the closed boundary fracturing horizontal well to generate fracturing horizontal well data, and to pre-process the fracturing horizontal well data to obtain standard fracturing horizontal well data.

[0065] A boundary fracturing oil and gas simulation module is configured to identify the closed boundary fracturing characteristics of the standard fracturing horizontal well data to obtain closed boundary fracturing data, to extract the fracture characteristics of the closed boundary fracturing data to generate boundary fracture characteristic information, to determine the fracturing porosity of the standard fracturing horizontal well data according to the boundary fracture characteristic information to obtain fracturing porosity data, and to simulate the oil and gas characteristics of the standard fracturing horizontal well data based on the boundary fracture characteristic information and the fracturing porosity data to generate oil and gas characteristic data.

[0066] A fracturing horizontal well bottom hole pressure measurement module is configured to extract the fracture parameters of the boundary fracture characteristic information through the oil and gas characteristic data to obtain boundary fracture parameter information, to identify the wellbore position of the standard fracturing horizontal well data according to the boundary fracture parameter information to obtain wellbore position data, to associate and map the boundary fracture parameter information and the wellbore position data to generate fracture wellbore associated data, and to measure the bottom hole pressure of the standard fracturing horizontal well data based on the fracture wellbore associated data to obtain bottom hole pressure data.

[0067] The oil and gas unsteady productivity index measurement module is configured to perform boundary fracture mutual coupling analysis on boundary fracture characteristic information according to well bottom pressure data, to generate boundary fracture relationship data; to perform oil and gas flow state recognition on oil and gas characteristic data according to the boundary fracture relationship data, to obtain oil and gas flow state data; to perform oil and gas unsteady productivity index measurement on standard fracturing horizontal well data through the oil and gas flow state data, to generate oil and gas unsteady productivity index; to perform unsteady productivity prediction model construction using the oil and gas unsteady productivity index, to obtain an unsteady productivity prediction model; and to perform unsteady productivity prediction on the standard fracturing horizontal well data based on the unsteady productivity prediction model, to generate an unsteady productivity prediction report.

[0068] The present application can provide basic information for subsequent analysis by using the sensor to collect data of the closed boundary fracturing horizontal well through the data collection module, ensuring to obtain original and direct information about the fracturing horizontal well; the data of the fracturing horizontal well is preprocessed, which is helpful to eliminate noise and abnormal values in the data, and can ensure the consistency and accuracy of the data, providing key data input for subsequent fracturing porosity determination and oil and gas feature simulation. The closed boundary fracturing feature of the standard fracturing horizontal well data can be recognized through the closed boundary fracturing oil and gas simulation module, which can evaluate the key of the fracture network and the reservoir connectivity; the fracture feature of the closed boundary fracturing data can be extracted, which can clearly determine the boundary fracture feature information; the fracturing porosity of the standard fracturing horizontal well data can be determined according to the boundary fracture feature information, which can accurately reflect the pore structure of the fracturing area, and is crucial for evaluating the reservoir characteristics of the oil and gas reservoir; the oil and gas feature simulation of the standard fracturing horizontal well data can be carried out based on the boundary fracture feature information and the fracturing porosity data, which can simulate the distribution and flow of oil and gas in the fracturing horizontal well. The fracture parameter of the boundary fracture feature information can be extracted through the oil and gas feature data by the fracturing horizontal well bottom pressure determination module, which can describe the geometric characteristics and distribution characteristics of the fracture in detail; the wellbore position of the standard fracturing horizontal well data can be recognized according to the boundary fracture parameter information, which can determine the accurate position of the wellbore in the oil and gas reservoir; the boundary fracture parameter information and the wellbore position data can be associated and mapped, which can associate the fracture parameter with the wellbore position, providing data support for the bottom hole pressure determination; the bottom hole pressure data can be determined based on the fracture wellbore associated data of the standard fracturing horizontal well data. The boundary fracture mutual coupling analysis of the boundary fracture feature information can be carried out according to the bottom hole pressure data through the oil and gas unsteady state productivity index measurement module, which can reveal the interaction and influence between the fractures; the oil and gas flow state of the oil and gas feature data can be recognized according to the boundary fracture relationship data, which can identify the flow state of oil and gas in the fracture network; the oil and gas unsteady state productivity index of the standard fracturing horizontal well data can be measured through the oil and gas flow state data, which can provide dynamic change information of the oil and gas production; the unsteady state productivity prediction model is constructed by using the oil and gas unsteady state productivity index, and the unsteady state productivity prediction model is obtained; the unsteady state productivity of the standard fracturing horizontal well data is predicted based on the unsteady state productivity prediction model, and the unsteady state productivity prediction report is generated. Therefore, the fracture mutual relationship between the bottom hole pressure and the boundary fracture feature is identified by using the data processing technology, the pattern recognition technology and the deep learning technology, and the oil and gas flow state of the oil and gas feature is detected according to the boundary fracture relationship, so that the accuracy of the unsteady state productivity prediction is improved. BRIEF DESCRIPTION OF DRAWINGS

[0069] Figure 1 It is a step flow chart for the closed boundary fracturing horizontal well unsteady state productivity prediction method;

[0070] Figure 2 For Figure 1 Detailed implementation step flow diagram of step S3 in the method;

[0071] Figure 3 For Figure 2 Detailed implementation step flow diagram of step S34 in the method;

[0072] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0073] The technical method of the present application will be described clearly and completely below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0074] In addition, the accompanying drawings are only schematic illustrations of the present application, and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus repeated description thereof will be omitted. Some block diagrams shown in the drawings are functional entities, which do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in the form of software, or in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0075] It should be understood that although the terms "first", "second" and the like can be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, without departing from the scope of the exemplary embodiments, a first element can be called a second element, and similarly a second element can be called a first element. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0076] To achieve the above-mentioned purpose, please refer to Figures 1 to 3 , the method for predicting the unsteady productivity of a closed boundary fracturing horizontal well, comprising the following steps:

[0077] Step S1: data acquisition is performed on the closed boundary fracturing horizontal well to generate fracturing horizontal well data; the fracturing horizontal well data is preprocessed to obtain standard fracturing horizontal well data;

[0078] Step S2: closed boundary fracturing feature recognition is performed on the standard fracturing horizontal well data to obtain closed boundary fracturing data; fracture feature extraction is performed on the closed boundary fracturing data to generate boundary fracture feature information; fracturing porosity determination is performed on the standard fracturing horizontal well data according to the boundary fracture feature information to obtain fracturing porosity data; and oil and gas feature simulation is performed on the standard fracturing horizontal well data based on the boundary fracture feature information and the fracturing porosity data to generate oil and gas feature data;

[0079] Step S3: fracture parameter information is obtained by extracting the fracture parameters of the boundary fracture feature information through the oil and gas feature data; wellbore position data is obtained by recognizing the wellbore position of the standard fracturing horizontal well data according to the boundary fracture parameter information; and fracture wellbore correlation data is generated by correlating and mapping the boundary fracture parameter information and the wellbore position data; and wellbore pressure data is obtained by determining the wellbore pressure of the standard fracturing horizontal well data based on the fracture wellbore correlation data;

[0080] Step S4: boundary fracture relationship data is generated by performing boundary fracture mutual coupling analysis on the boundary fracture feature information according to the wellbore pressure data; oil and gas flow state data is obtained by recognizing the oil and gas flow state of the oil and gas feature data according to the boundary fracture relationship data; oil and gas non-steady-state productivity index is generated by measuring and calculating the oil and gas non-steady-state productivity index of the standard fracturing horizontal well data through the oil and gas flow state data; a non-steady-state productivity prediction model is constructed by using the oil and gas non-steady-state productivity index to obtain a non-steady-state productivity prediction model; and a non-steady-state productivity prediction report is generated by performing non-steady-state productivity prediction on the standard fracturing horizontal well data based on the non-steady-state productivity prediction model.

[0081] The application uses a sensor to collect data of a closed boundary fracturing horizontal well, ensures to obtain original and direct information about the fracturing horizontal well, can provide basic information for subsequent analysis, pre-processes the fracturing horizontal well data, helps to eliminate noise and abnormal values in the data, can ensure consistency and accuracy of the data, and provides key data input for subsequent fracturing porosity determination and oil and gas feature simulation. The closed boundary fracturing feature identification of the standard fracturing horizontal well data can evaluate key of the fracture network and reservoir connectivity. The fracture feature extraction of the closed boundary fracturing data can clearly determine the boundary fracture feature information. The fracturing porosity determination of the standard fracturing horizontal well data according to the boundary fracture feature information can accurately reflect the pore structure of the fracturing area, which is crucial for evaluating the reservoir characteristics of the oil and gas reservoir. The oil and gas feature simulation of the standard fracturing horizontal well data based on the boundary fracture feature information and the fracturing porosity data can simulate the distribution and flow of oil and gas in the fracturing horizontal well. The fracture parameter extraction of the boundary fracture feature information through the oil and gas feature data can describe the geometric characteristics and distribution characteristics of the fracture in detail. The wellbore position identification of the standard fracturing horizontal well data according to the boundary fracture parameter information can determine the accurate position of the wellbore in the oil and gas reservoir. The correlation mapping of the boundary fracture parameter information and the wellbore position data can correlate the fracture parameters and the wellbore position, and provide data support for the bottom hole pressure determination. The bottom hole pressure determination of the standard fracturing horizontal well data based on the fracture wellbore correlation data can clearly determine the bottom hole pressure data. The boundary fracture mutual coupling analysis of the boundary fracture feature information according to the bottom hole pressure data can reveal the interaction and influence between the fractures. The oil and gas flow state identification of the oil and gas feature data according to the boundary fracture relationship data can identify the flow state of oil and gas in the fracture network. The oil and gas non-steady state productivity index measurement of the standard fracturing horizontal well data through the oil and gas flow state data can provide dynamic change information of the oil and gas production. The non-steady state productivity prediction model construction using the oil and gas non-steady state productivity index can obtain a non-steady state productivity prediction model. The non-steady state productivity prediction of the standard fracturing horizontal well data based on the non-steady state productivity prediction model can generate a non-steady state productivity prediction report. Therefore, through the data processing technology, the pattern recognition technology, and the deep learning technology, the application realizes the fracture mutual relationship identification of the bottom hole pressure and the boundary fracture feature, and realizes the oil and gas flow state detection of the oil and gas feature according to the boundary fracture relationship, so as to improve the accuracy of the non-steady state productivity prediction.

[0082] In the embodiment of the application, reference Figure 1 The is a step flow diagram of the closed boundary fracturing horizontal well non-steady state productivity prediction method of the application. In the example, the closed boundary fracturing horizontal well non-steady state productivity prediction method includes the following steps:

[0083] Step S1: data acquisition is performed on the closed boundary fracturing horizontal well to generate fracturing horizontal well data; the fracturing horizontal well data is preprocessed to obtain standard fracturing horizontal well data;

[0084] In the embodiment of the present application, by deploying high-precision sensors such as pressure sensors and temperature sensors, the physical parameters in the fracturing horizontal well are monitored and recorded in real time; the pressure change, fluid flow and temperature distribution in the fracturing process are specifically captured to generate fracturing horizontal well data; the fracturing horizontal well data is preprocessed, and the preprocessing steps include removing unique attributes, processing missing values, attribute encoding, data standardization and regularization; for example, for missing values, the mean interpolation or similar mean interpolation method is used; for data standardization, the min-max standardization method is used to scale the data to a specified range to eliminate the influence of different orders of magnitude attributes; so as to obtain standard fracturing horizontal well data.

[0085] Step S2: closed boundary fracturing feature recognition is performed on the standard fracturing horizontal well data to obtain closed boundary fracturing data; fracture feature information is generated by extracting the fracture features of the closed boundary fracturing data; the fracturing porosity data is obtained by determining the fracturing porosity of the standard fracturing horizontal well data according to the boundary fracture feature information; and the oil and gas feature data is generated by simulating the oil and gas features of the standard fracturing horizontal well data based on the boundary fracture feature information and the fracturing porosity data;

[0086] In the embodiment of the present application, the feature recognition technology is used to perform closed boundary fracturing feature recognition on the standard fracturing horizontal well data, and the boundary fracturing feature parameters and the boundary fracturing situation are specifically identified; the data detection technology such as edge detection algorithm is used to perform fracture feature recognition on the closed boundary fracturing data, and the boundary fracture edge and width information is specifically obtained to obtain the boundary fracture feature information; the microseismic monitoring technology is used to perform fracture porosity recognition on the standard fracturing horizontal well data according to the boundary fracture feature information, and the fracture porosity volume is calculated to obtain the fracture porosity volume; the reservoir porosity is calculated by the fracture porosity volume to obtain the fracturing porosity data; based on the fracture feature information and the determined fracturing porosity data, the physical simulation experiment technology such as visual three-dimensional physical simulation experiment technology is used to simulate the oil and gas flow process; specifically, the distribution and flow behavior of oil and gas in the reservoir are observed by physical simulation experiments such as two-dimensional sand box simulation experiment or three-dimensional consolidation model experiment to generate oil and gas feature data.

[0087] Step S3: crack parameter extraction is performed on the boundary crack feature information through the oil and gas feature data to obtain boundary crack parameter information; wellbore position data is obtained by identifying the wellbore position of the standard fracturing horizontal well data according to the boundary crack parameter information; crack wellbore correlation data is generated by correlating and mapping the boundary crack parameter information and the wellbore position data; and the bottom hole pressure data is obtained by determining the bottom hole pressure of the standard fracturing horizontal well data based on the crack wellbore correlation data.

[0088] In the embodiment of the present application, the image-based crack segmentation technology is used, for example, the threshold segmentation method using OpenCV library is used for crack segmentation; the orthogonal skeleton line method is applied to the segmented crack image to identify the crack skeleton line; and the crack width is determined using the width calculation method to obtain the boundary crack parameter information; the range of the wellbore is calculated according to the boundary crack parameter information based on the hierarchical parameterization method to obtain the wellbore range calculation data; the wellbore position data is obtained by identifying the wellbore position according to the wellbore range calculation data; the crack wellbore correlation data is generated by correlating and mapping the boundary crack parameter information and the wellbore position data, specifically by integrating the crack parameter and the wellbore position information using the data fusion technology; and the bottom hole pressure data is obtained by real-time monitoring and dynamic pressure recording of the standard fracturing horizontal well data using the pressure sensor based on the crack wellbore correlation data.

[0089] Step S4: boundary crack relationship data is generated by performing boundary crack mutual coupling analysis on the boundary crack feature information according to the bottom hole pressure data; oil and gas flow state data is obtained by identifying the oil and gas flow state of the oil and gas feature data according to the boundary crack relationship data; oil and gas non-steady-state productivity index is generated by performing oil and gas non-steady-state productivity index calculation on the standard fracturing horizontal well data through the oil and gas flow state data; non-steady-state productivity prediction model is obtained by constructing the non-steady-state productivity prediction model using the oil and gas non-steady-state productivity index; and non-steady-state productivity prediction report is generated by performing non-steady-state productivity prediction on the standard fracturing horizontal well data based on the non-steady-state productivity prediction model.

[0090] In the embodiment of the present application, the mutual correlation of the boundary fracture characteristic information is determined according to the bottom hole pressure data, and the mutual constraint and mutual cooperation between the boundary fractures are specifically obtained, so as to obtain the mutual correlation between the boundary fractures, identify the coupling relationship of the mutual correlation between the boundary fractures, and generate the coupling relationship; the mutual correlation between the boundary fractures and the coupling relationship are integrated by using the data fusion technology, so as to obtain the boundary fracture relationship data; the oil and gas flow characteristic mark is marked on the oil and gas characteristic data according to the boundary fracture relationship data, and the characteristic enhancement is performed on the oil and gas flow characteristic mark; the oil and gas flow state is identified according to the oil and gas flow mark characteristic enhancement data by using the pattern recognition technology, so as to obtain the oil and gas flow state data; the non-steady state productivity of the equivalent well is determined through the mirror inversion and pressure drop superposition principle according to the boundary fracture relationship data and the oil and gas flow state data; a plurality of fracture parameters, time steps and corresponding non-steady state productivities are specifically calculated, and the oil and gas non-steady state productivity index is generated; the non-steady state productivity prediction model is constructed by using the oil and gas non-steady state productivity index, and the model is constructed by using the machine learning algorithm, so as to obtain the non-steady state productivity prediction model; the non-steady state productivity of the standard fracturing horizontal well data is predicted based on the non-steady state productivity prediction model; the well productivity under the non-steady state condition is predicted by the model, and the non-steady state productivity prediction report is generated.

[0091] Preferably, the step S2 comprises the following steps:

[0092] Step S21: performing non-linear feature decomposition on the standard fracturing horizontal well data to generate non-linear fracturing horizontal well features; performing fracturing feature enhancement on the non-linear fracturing horizontal well features to obtain fracturing feature enhancement data; and performing closed boundary fracturing feature extraction on the fracturing feature enhancement data to obtain closed boundary fracturing data.

[0093] Step S22: performing fracturing morphology feature extraction on the closed boundary fracturing data to generate fracturing morphology feature data; performing geometric feature quantization on the standard fracturing horizontal well data to obtain fracturing morphology geometric features; and performing fracture feature identification on the fracturing morphology geometric features to generate boundary fracture characteristic information.

[0094] Step S23: determining the distribution range of the boundary fracture characteristic information to obtain the boundary fracture distribution range; performing fracture spatial distribution feature identification on the boundary fracture distribution range to generate fracture spatial distribution features; and performing fracture spatial distribution morphology feature on the fracturing morphology geometric features according to the fracture spatial distribution features to obtain the fracture spatial distribution morphology features.

[0095] Step S24: crack density calculation is performed on the crack spatial distribution pattern characteristics to generate crack spatial density data; based on the crack spatial density data, fracture pore characteristics are obtained by performing fracture pore feature on the boundary crack feature information; and by performing fracture porosity measurement on the standard fracturing horizontal well data through the fracture pore characteristics, fracture porosity data are obtained;

[0096] Step S25: porosity gradient division is performed on the fracture porosity data to generate porosity gradient data; and based on the porosity gradient data and the boundary crack feature information, oil and gas feature simulation is performed on the standard fracturing horizontal well data to generate oil and gas feature data.

[0097] In the embodiment of the present application, a singular spectrum analysis (SSA) method with adaptive threshold is used to perform nonlinear feature decomposition on standard fracturing horizontal well data, to generate nonlinear fracturing horizontal well features; the SSA method separates periodic components and trend components in the data by constructing data reconstruction of different time scales, to identify nonlinear features; a feature enhancement technique based on wavelet transform is used to perform feature enhancement processing on the nonlinear fracturing horizontal well features, to specifically provide feature information in a time-frequency domain, to highlight feature frequencies generated by fracturing events by appropriately selecting a wavelet basis and a decomposition level, to enhance fracturing features; a feature extraction technique is applied to extract closed boundary fracturing features from the fracturing feature enhanced data, to specifically extract geometric and texture information of fractures; an image processing technique, such as a fracture segmentation method based on a histogram, is used to process the closed boundary fracturing data, to extract morphological features of fractures; a three-dimensional reconstruction technique is used to quantize geometric features of standard fracturing horizontal well data, to specifically quantize geometric information of morphologies and spatial structures of fractures; a feature recognition technique is used to recognize fracturing morphological geometric features, to specifically recognize boundary fracture feature information. A fracture network analysis method based on fractal theory is used to determine the boundary fracture feature information, to specifically determine a distribution range of fractures; complexity of a fracture network is quantized, and a fractal dimension of the fracture network is calculated, to determine the distribution range and density of the fractures; a distribution characteristic space of the boundary fracture distribution range is simulated according to spatial distribution characteristics of the fractures; a fracture skeleton line extraction method based on width calculation, such as an orthogonal skeleton line method, is used to extract the fracture skeleton line from the fracturing morphological geometric features, and the width and length of the fracture are calculated, to obtain spatial distribution morphological features of the fractures; a fracture density calculation is performed by using a numerical simulation of a fractured oil and gas reservoir, to specifically analyze the density of the spatial distribution morphological features of the fractures; based on the fracture spatial density data, geometric features of the fractures and flow characteristics of fluid are calculated, to obtain porosity of a reservoir, to obtain fracturing porosity features; fracturing porosity data is obtained by performing fracturing porosity measurement on the standard fracturing horizontal well data according to the fracturing porosity features; a width calculation method based on an image, such as the orthogonal skeleton line method, is used to perform gradient division on the porosity data according to the fracturing porosity data, to generate porosity gradient data; different porosity gradient regions are divided by analyzing spatial changes of the porosity; oil and gas feature data is generated by simulating oil and gas features of the standard fracturing horizontal well data by using the porosity gradient data and the boundary fracture feature information.

[0098] As an example of the present application, reference is made to Figure 2 In this example, the step S3 comprises:

[0099] Step S31: fracture geometric feature recognition is performed on the boundary fracture feature information, to generate fracture geometric feature data; fracture geometric parameters are extracted from the fracture geometric feature data according to the oil and gas feature data, to obtain the fracture geometric parameters;

[0100] Step S32: crack connectivity detection is performed on the boundary crack characteristic information through the crack geometric parameters to generate crack connectivity data; crack permeability evaluation is performed on the boundary crack characteristic information based on the crack connectivity data to obtain crack permeability data; crack fluid flow simulation is performed on the oil and gas characteristic data according to the crack permeability data to obtain crack fluid flow simulation data;

[0101] Step S33: the boundary crack parameter information is obtained by integrating the crack connectivity data, the crack permeability data and the crack fluid flow simulation data; wellbore characteristic identification is performed on the standard fracturing horizontal well data to generate wellbore characteristic data; wellbore position detection is performed on the wellbore characteristic data according to the boundary crack parameter information to obtain wellbore position data.

[0102] Step S34: crack wellbore correlation data is generated by correlating and mapping the boundary crack parameter information and the wellbore position data; wellbore pressure measurement is performed on the standard fracturing horizontal well data based on the crack wellbore correlation data to obtain wellbore pressure data.

[0103] In the embodiment of the present application, the geometric feature recognition method is adopted, the crack shape is specifically extracted, and the crack parameter information is obtained through image processing techniques such as connected domain denoising, fracture connection, edge detection and crack skeletonization, so as to generate crack geometric characteristic data; the crack geometric data is obtained by extracting the width, length and other geometric parameters of the crack through the orthogonal skeleton line method technology based on the analysis of the crack geometric characteristic data according to the oil and gas characteristic data; the crack connectivity data is generated by performing crack connectivity detection on the boundary crack characteristic information through the crack geometric parameters by using the multi-scale crack network distribution inversion method in the crack medium based on the hierarchical parameterization and data-driven evolutionary optimization method; the crack permeability data is obtained by performing crack permeability evaluation on the boundary crack characteristic information based on the crack connectivity data by using the fluid flow permeability evaluation technology; the crack fluid flow simulation data is obtained by performing crack fluid flow simulation on the oil and gas characteristic data according to the crack permeability data by using the fluid flow simulation technology; the boundary crack parameter information is obtained by integrating the crack connectivity data, the crack permeability data and the crack fluid flow simulation data by using the data fusion technology; the wellbore characteristic data is generated by performing wellbore characteristic identification on the standard fracturing horizontal well data by using the feature recognition technology; the wellbore position data is obtained by performing wellbore position detection according to the boundary crack parameter information in combination with the wellbore characteristic data, specifically detecting the position information of the wellbore position; the boundary crack parameter information is correlated and mapped with the wellbore position data by using the data correlation technology, specifically correlating the boundary crack with the wellbore position; the wellbore pressure data is obtained by performing wellbore pressure measurement on the standard fracturing horizontal well data based on the crack wellbore correlation data by using the real-time pressure monitoring technology.

[0104] As an example of the present application, reference is made toFigure 3 As shown, the step S34 includes, in the present example:

[0105] Step S341: Connectivity quantification is performed on the fracture connectivity data to obtain fracture connectivity quantitative data; the fracture connectivity quantitative data is associated with the wellbore position data to generate fracture wellbore connectivity association data;

[0106] Step S342: Permeability parameterization is performed on the fracture permeability data to obtain fracture permeability parameterized data; the fracture permeability parameterized data is associated with the wellbore position data to generate fracture wellbore permeability association data;

[0107] Step S343: Flow characteristic extraction is performed on the fracture fluid flow simulation data to obtain fracture fluid flow characteristic data; the fracture fluid flow characteristic data is associated with the wellbore position data to generate fracture wellbore flow property association data;

[0108] Step S344: The fracture wellbore connectivity association data, the fracture wellbore permeability association data, and the fracture wellbore flow property association data are integrated to obtain fracture wellbore association data;

[0109] Step S345: Based on the fracture wellbore association data, bottom hole pressure determination is performed on the standard fracturing horizontal well data to obtain bottom hole pressure data.

[0110] In the embodiment of the present application, the fracture connectivity data is quantified, specifically by calculating the hydraulic conductivity of the fracture and the shortest distance between the fracture clusters, specifically quantifying the fracture connectivity, so as to obtain the fracture connectivity quantitative data; the fracture connectivity quantitative data is associated with the wellbore position data, and the influence degree of the fracture connectivity on the wellbore position is specifically calculated, so as to generate the fracture wellbore connectivity association data; the fracture permeability parameterization data is obtained by using the fracture permeability evaluation technology and parameterizing the fracture permeability data; the fracture permeability parameterization data is associated with the wellbore position data, and the reservoir fluid flow dominated by the fracture is simulated by a numerical simulation method, so as to generate the fracture wellbore permeability association data; the fracture fluid flow characteristic data is obtained by using the feature extraction technology and extracting the flow characteristics of the fracture fluid flow simulation data; the fracture fluid flow characteristic data is associated with the wellbore position data, and the fracture network influence on the fluid flow is analyzed, so as to generate the fracture wellbore flowability association data; the fracture wellbore association data is obtained by using the data fusion technology and integrating the fracture wellbore connectivity association data, the fracture wellbore permeability association data and the fracture wellbore flowability association data; and the wellbore pressure data is obtained by using the underbalanced drilling bottom hole pressure measurement technology, the while-drilling measurement and the annulus multiphase flow bottom hole pressure calculation model, and determining the bottom hole pressure of the standard fracturing horizontal well data based on the fracture wellbore association data.

[0111] Preferably, the step S345 includes the following steps:

[0112] Step S3451: fracture wellbore association data is used to extract the bottom hole stress characteristics of the standard fracturing horizontal well data, so as to generate the bottom hole stress characteristics; and the bottom hole stress distribution simulation of the standard fracturing horizontal well data is simulated according to the bottom hole stress characteristics, so as to obtain the bottom hole pressure distribution simulation data;

[0113] Step S3452: the bottom hole pressure distribution simulation data is used to identify the bottom hole pressure response, so as to generate the bottom hole pressure response data; the bottom hole pressure response data is used to calculate the pressure numerical value, so as to obtain the bottom hole pressure numerical information; and the bottom hole pressure numerical information is used to divide the pressure gradient, so as to generate the bottom hole pressure gradient data;

[0114] Step S3453: the bottom hole pressure distribution simulation data is monitored in real time by using the bottom hole pressure gradient data, so as to obtain the bottom hole pressure change data; and the bottom hole pressure state information is generated by using the bottom hole pressure change data to identify the bottom hole pressure state;

[0115] Step S3454: performing bottom hole pressure anomaly detection on the bottom hole pressure state information to obtain bottom hole pressure anomaly data; performing pressure stability evaluation on the bottom hole pressure distribution simulation data according to the bottom hole pressure anomaly data to obtain bottom hole pressure stability data; and performing bottom hole pressure measurement on the standard fracturing horizontal well data based on the bottom hole pressure stability data to obtain bottom hole pressure data.

[0116] In the embodiment of the present application, based on the fracture wellbore correlation data, the downhole acoustic detection technology is used to extract the bottom hole stress characteristics of the standard fracturing horizontal well data; the downhole acoustic reflection signal is specifically analyzed to determine the stress state of the bottom hole rock and the development of the fracture, and the bottom hole stress characteristics are generated; according to the bottom hole stress characteristics, the bottom hole pressure distribution simulation of the standard fracturing horizontal well data is performed using the managed pressure drilling pressure detection technology, and the bottom hole pressure distribution simulation data is obtained; the real-time formation pressure inversion method based on back pressure excitation response is used to perform bottom hole pressure response identification on the bottom hole pressure distribution simulation data, and the bottom hole pressure response data is generated; the method inverts the formation pressure by applying back pressure excitation at the wellhead, combining the annulus data calculated by the simplified two-phase flow model and the ground measurement data; the pressure response numerical value is extracted from the bottom hole pressure response data, and the pressure numerical value is calculated to obtain the bottom hole pressure numerical value information; the pressure gradient is divided according to the bottom hole pressure numerical value information, and the pressure gradient is specifically divided into three levels, so that the bottom hole pressure gradient data is obtained; the real-time bottom hole pressure monitoring of the bottom hole pressure distribution simulation data is specifically performed using the permanent optical and electrical composite cable monitoring technology through the bottom hole pressure gradient data, and the bottom hole pressure change data is obtained; the bottom hole pressure state identification is performed on the bottom hole pressure change data, and the bottom hole pressure state mode such as high pressure state and low pressure state is specifically identified, so that the bottom hole pressure state information is obtained; the real-time formation pressure inversion method based on back pressure excitation response is used to perform bottom hole pressure anomaly detection on the bottom hole pressure state information, and the abnormal state of the bottom hole pressure being too high or too low is specifically detected to obtain the bottom hole pressure anomaly data; the pressure stability evaluation is performed on the bottom hole pressure distribution simulation data according to the bottom hole pressure anomaly data to obtain the bottom hole pressure stability data; and the bottom hole pressure measurement is performed on the standard fracturing horizontal well data based on the bottom hole pressure stability data using the bottom hole pressure measurement technology, and the bottom hole pressure data is obtained.

[0117] Preferably, step S4 comprises the following steps:

[0118] Step S41: performing boundary fracture mutual coupling analysis on the boundary fracture characteristic information according to the bottom hole pressure data to generate boundary fracture relationship data;

[0119] Step S42: performing oil and gas flow state identification on the oil and gas characteristic data according to the boundary fracture relationship data to obtain oil and gas flow state data;

[0120] Step S43: Calculate the oil and gas unsteady-state productivity index by the oil and gas flow state data and the standard fracturing horizontal well data to generate the oil and gas unsteady-state productivity index;

[0121] Step S44: Construct the unsteady-state productivity prediction model according to the oil and gas unsteady-state productivity index to obtain the unsteady-state productivity prediction model; and train the unsteady-state productivity prediction model by the oil and gas unsteady-state productivity index to obtain the unsteady-state productivity prediction training model;

[0122] Step S45: Perform model cross-validation evaluation on the unsteady-state productivity prediction training model to obtain model evaluation data; and adjust the model parameters of the unsteady-state productivity prediction training model by the model evaluation data to obtain the unsteady-state productivity prediction model;

[0123] Step S46: Perform unsteady-state productivity prediction on the standard fracturing horizontal well data based on the unsteady-state productivity prediction model to generate an unsteady-state productivity prediction report.

[0124] In the embodiment of the present application, the bottom hole pressure data is monitored in real time by using the distributed acoustic sensing technology (DAS), the mutual influence and coupling relationship between the cracks are analyzed in detail, and the boundary crack relationship data is generated; the optical signal is converted into the seismic vibration signal by using the machine learning algorithm such as the heterodyne demodulation algorithm, and then the signal is denoised to obtain the crack network information and generate the boundary crack relationship data; the oil and gas characteristic data is analyzed in detail by using the physical-guided semi-supervised learning fluid prediction and reservoir identification method, the oil and gas flow state is identified, and the oil and gas flow state data is obtained; the real-time dynamic monitoring of the whole wellbore pressure is performed in detail by using the permanent optical-electric composite cable monitoring new technology, the oil and gas flow state is identified, and the oil and gas flow state data is obtained; the productivity analysis technology under the unsteady state condition of the vertical fracturing horizontal well is applied, the pressure drop formula is derived, and the relationship between the wellbore and the reservoir coupling is established; the seepage characteristic analysis technology is used, the index of the bottom hole pressure is obtained by the pressure drop superposition technology, and the unsteady state oil and gas productivity index is generated; the unsteady state productivity prediction model is constructed according to the unsteady state oil and gas productivity index; specifically, the infinite conductivity model and the finite conductivity model in the productivity analysis under the unsteady state condition of the vertical fracturing horizontal well are used; based on the Green function and the Newman product principle, the pressure drop formula of the vertical fracturing horizontal well under the unsteady state condition is derived when the vertical fracturing horizontal well is produced, and the unsteady state productivity prediction model is established; the point source solution theory, Pedrosa transformation, perturbation transformation, Laplace transformation, finite cosine transformation, numerical discretization and superposition principle in the unsteady state productivity prediction model are used to solve the established mathematical model; the typical production decline curve is drawn by using the Stehfest numerical inversion technology, and the curve characteristics are divided into different flow stages, and the model verification and comparison are performed at the same time; the model cross-validation evaluation of the unsteady state productivity prediction training model is performed; the K-fold cross-validation method is used to divide the data set into K subsets, each subset is used as the test set in turn, the remaining subsets are used as the training set, the performance index of the model on each subset is calculated, and the model evaluation data is obtained; the model parameter adjustment of the unsteady state productivity prediction training model is performed through the model evaluation data. The adjustment parameters include the permeability modulus, the branch well length and the number of branch wells and other sensitive parameters, so as to summarize the influence of different parameters on the production performance; with the increase of the stress sensitivity coefficient, the oil production and the cumulative oil production of the oil well decrease; or when the total number of cracks is small, the segment cluster ratio of the cracks has a greater influence on the cumulative oil production, under the same number of cracks, the greater the segment cluster ratio, the greater the cumulative oil production; the unsteady state productivity prediction of the standard fracturing horizontal well data is performed based on the unsteady state productivity prediction model, and the unsteady state productivity prediction report is generated; the new bottom hole pressure data is input into the trained model, and the predicted unsteady state productivity index is output, and the unsteady state productivity prediction report is generated.

[0125] Preferably, the step S41 comprises the following steps:

[0126] Step S411: pressure distribution feature extraction is performed on the bottom hole pressure data to obtain bottom hole pressure distribution features; bottom hole pressure peak region identification is performed on the bottom hole pressure distribution features to generate bottom hole pressure peak regions;

[0127] Step S412: boundary fracture feature information is identified according to the fracture spatial density data to obtain boundary fracture concentrated regions; the boundary fracture concentrated regions are mapped and matched with the bottom hole pressure peak regions to generate fracture pressure mapping data;

[0128] Step S413: fracture pressure influence degree calculation is performed on the bottom hole pressure data through the fracture pressure mapping data to obtain fracture pressure influence data; fracture sensitivity determination is performed on the boundary fracture feature information according to the fracture pressure influence data to generate fracture sensitivity data;

[0129] Step S414: boundary fracture constraint feature extraction is performed on the fracture sensitivity data using the fracture connectivity data to obtain boundary fracture constraint action data; boundary fracture permeation synergy feature extraction is performed on the boundary fracture constraint action data through the fracture permeability data to generate boundary fracture synergy action data; boundary fracture mutual coupling identification is performed on the boundary fracture feature information according to the boundary fracture constraint action data and the boundary fracture synergy action data to obtain boundary fracture relationship data.

[0130] In the embodiment of the present application, based on the adaptive threshold technology, the bottom hole pressure data is processed in the form of image, the optimal threshold is automatically obtained through the adaptive threshold method, the extraction of the bottom hole pressure distribution characteristics is realized, and the bottom hole pressure distribution characteristics are obtained; the bottom hole pressure distribution characteristics are analyzed by using the image processing technology and combining the directionality and continuity characteristics of the fracture image, the pressure peak area is identified, and the bottom hole pressure peak area is generated; according to the fracture space density data, the irregular micro-fracture network quantitative characterization method is adopted, the seepage capacity of the porous medium is calculated, the fracture concentration area is identified according to the boundary fracture characteristic information, and the boundary fracture concentration area is obtained; the boundary fracture concentration area and the bottom hole pressure peak area are mapped and matched, the fracture pressure mapping data are obtained through the numerical simulation method such as the finite difference or the computational fluid dynamics method (CFD); the fracture pressure influence degree of the bottom hole pressure data is calculated through the fracture pressure mapping data, and the fracture pressure influence data are obtained; according to the fracture pressure influence data, the fracture sensitivity determination technology is used to determine the fracture sensitivity of the boundary fracture characteristic information, and the fracture sensitivity data are obtained; the boundary fracture constraint characteristic extraction of the fracture sensitivity data is performed on the fracture sensitivity data by using the structure fracture depth inversion technology according to the fracture connectivity data, and the boundary fracture constraint action data are obtained; the boundary fracture permeation synergy characteristic extraction of the boundary fracture constraint action data is performed by combining the synergy analysis technology according to the fracture permeability data, and the boundary fracture synergy action data are obtained; according to the boundary fracture constraint action data and the boundary fracture synergy action data, the boundary fracture mutual coupling of the boundary fracture characteristic information is identified by using the productivity analysis technology, and the boundary fracture relationship data are obtained.

[0131] Preferably, the step S42 comprises the following steps:

[0132] Step S421: according to the boundary fracture relationship data, the fracture-oil and gas flow action corresponding of the oil and gas characteristic data is performed, and the fracture-oil and gas flow action data are obtained;

[0133] Step S422: the oil and gas flow characteristic marking of the oil and gas characteristic data is performed, and the oil and gas flow marking data are generated; the flow unit division of the oil and gas characteristic data is performed through the oil and gas flow marking data, and the flow unit division data are obtained; the flow characteristic extraction of the flow unit division data is performed, and the oil and gas flow characteristic data are generated;

[0134] Step S423: based on the fracture-oil and gas flow action data, the fracture oil and gas flow network construction of the oil and gas flow characteristic data is performed, and the fracture oil and gas flow network is obtained; the oil and gas flow direction path recognition of the fracture oil and gas flow network is performed, and the oil and gas flow direction path data are generated; according to the oil and gas flow direction path data, the oil and gas flow direction flow calculation of the fracture oil and gas flow network is performed, and the oil and gas flow direction flow data are obtained;

[0135] Step S424: Perform flow state anomaly detection on the fracture oil and gas flow network through the oil and gas flow direction flow data to generate oil and gas flow state anomaly data; perform flow state stability evaluation on the fracture oil and gas flow network to obtain flow state stability data;

[0136] Step S425: Integrate the oil and gas flow direction flow data, the oil and gas flow state anomaly data and the flow state stability data to obtain the oil and gas flow state data.

[0137] In the embodiment of the present application, according to the boundary fracture relationship data, the fracture-oil and gas flow action data is obtained by using the numerical simulation technology in the development of the fracture reservoir fluid flow numerical simulation research. Specifically, the fracture-oil and gas flow action data is obtained by identifying the influence of the fracture on the oil and gas flow through simulating the reservoir fluid flow dominated by the fracture network; the well bottom pressure peak area is identified by analyzing the well bottom pressure distribution characteristics, specifically by using the well bottom pressure technology; the well bottom pressure peak area is determined by analyzing the fluid flow state at the pore scale; the oil and gas flow feature data is marked by identifying and classifying the flow characteristics of the remaining oil, such as drop-shaped, column-shaped and film-shaped, etc.; the flow unit division data is obtained by dividing the flow units of the oil and gas feature data by using the fracture network simulation technology through the oil and gas flow marking data; the reservoir is divided into different flow units according to the connectivity of the fracture network and the fluid flow characteristics; the oil and gas flow characteristic data is extracted by using the oil and gas flow state analysis method by analyzing the flow unit division data; the oil and gas flow characteristic data is generated by specifically analyzing the fluid flow characteristics in each flow unit, such as flow rate, flow direction and flow path; the fracture oil and gas flow network is constructed by using the fracture oil and gas flow simulation technology based on the fracture-oil and gas flow action data, and the fracture oil and gas flow network is constructed by simulating the fluid flow in the fracture network. The oil and gas flow direction path data is obtained by identifying the oil and gas flow direction path by using the flow path analysis method by analyzing the fracture oil and gas flow network; the flow direction of the oil and gas is specifically determined by tracking the flow path of the fluid in the fracture network; the oil and gas flow direction flow rate is specifically calculated for the fracture oil and gas flow network by using the oil and gas flow direction simulation technology according to the oil and gas flow direction path data; the oil and gas flow direction flow rate data is obtained by simulating the flow rate on different flow direction paths; the flow state anomaly of the fracture oil and gas flow network is detected by using the flow state anomaly analysis method through the oil and gas flow direction flow rate data; the flow state anomaly is specifically identified by comparing the actual flow rate with the expected flow rate; the flow state stability data is obtained by using the stability analysis method by evaluating the stability of the flow state of the fracture oil and gas flow network; the stability of the flow state is evaluated by analyzing the stability of the fluid flow in the fracture network; the oil and gas flow state data is obtained by integrating the oil and gas flow direction flow rate data, the oil and gas flow state anomaly data and the flow state stability data by using the data fusion technology.

[0138] Preferably, the step S43 comprises the following steps:

[0139] Step S431: unit oil and gas volume data is obtained by performing unit oil and gas volume statistics on the oil and gas flow direction flow rate data; and the instantaneous oil and gas production data is generated by performing instantaneous oil and gas production calculation on the unit oil and gas volume data;

[0140] Step S432: Abnormal state type identification is performed on the oil and gas flow state abnormal data to obtain a flow abnormal state type; abnormal instantaneous production records are recorded on the instantaneous oil and gas production data according to the flow abnormal state type to generate abnormal state instantaneous production data; and the abnormal state instantaneous production data is accumulated to obtain abnormal state cumulative production data;

[0141] Step S433: Stable state instantaneous production records are recorded on the instantaneous oil and gas production data based on the flow state stability data to generate stable state instantaneous production data; and the stable state instantaneous production data is accumulated to obtain stable state cumulative production data;

[0142] Step S434: The stable state cumulative production data and the abnormal state cumulative production data are subjected to production value difference to obtain an oil and gas non-steady state productivity index.

[0143] In the embodiment of the present application, unit oil and gas volume statistics are performed on the oil and gas flow direction flow data, and the production calculation method in the pressure buildup test of the oil and water well test project is used; the oil production is calculated through the single well steady state flow method according to the dynamic data of the oil well fluid; the instantaneous oil and gas production is calculated through the unit oil and gas volume data, and the mapping relationship between various influencing factors and the natural gas well production is simulated to calculate the instantaneous oil and gas production data; the abnormal state type identification is performed on the oil and gas flow state abnormal data, and the abnormal state identification technology is used; the abnormal traffic state is quickly identified through the analysis of the moving object flow data to determine the flow abnormal state type; the abnormal instantaneous production records are recorded on the instantaneous oil and gas production data according to the flow abnormal state type to generate the abnormal state instantaneous production data; the abnormal state instantaneous production data is accumulated to obtain the abnormal state cumulative production data; the oil and gas production accumulation method is used to calculate the cumulative oil and gas production in the abnormal state by accumulating the instantaneous production in different time periods; the stable state instantaneous production records are recorded on the instantaneous oil and gas production data based on the flow state stability data to generate the stable state instantaneous production data; the fractured oil well production method in the oil and gas production calculation formula is adopted, and the formula is used to calculate the oil production based on the known fracturing fluid volume, inflow pressure and flow production pressure; the stable state instantaneous production data is accumulated to obtain the stable state cumulative production data; the oil and gas production accumulation method is used to calculate the cumulative oil and gas production in the stable state by accumulating the instantaneous production in different time periods; the stable state cumulative production data and the abnormal state cumulative production data are subjected to production value difference to obtain the oil and gas non-steady state productivity index; and the oil and gas production value difference calculation method is used to calculate the oil and gas non-steady state productivity index by performing numerical difference on the cumulative production data in the stable state and the abnormal state.

[0144] In the present specification, a closed boundary fracturing horizontal well non-steady state productivity prediction device is provided, comprising a prediction system for performing the closed boundary fracturing horizontal well non-steady state productivity prediction method described above, the prediction system comprising:

[0145] A data acquisition module is configured to acquire data of the closed boundary fracturing horizontal well to generate fracturing horizontal well data, and to preprocess the fracturing horizontal well data to obtain standard fracturing horizontal well data.

[0146] A boundary fracturing oil and gas simulation module is configured to identify closed boundary fracturing characteristics of the standard fracturing horizontal well data to obtain closed boundary fracturing data, to extract fracture characteristics of the closed boundary fracturing data to generate boundary fracture characteristic information, to determine fracturing porosity of the standard fracturing horizontal well data based on the boundary fracture characteristic information to obtain fracturing porosity data, and to simulate oil and gas characteristics of the standard fracturing horizontal well data based on the boundary fracture characteristic information and the fracturing porosity data to generate oil and gas characteristic data.

[0147] A fracturing horizontal well bottom hole pressure measurement module is configured to extract fracture parameters of the boundary fracture characteristic information based on the oil and gas characteristic data to obtain boundary fracture parameter information, to identify wellbore position of the standard fracturing horizontal well data based on the boundary fracture parameter information to obtain wellbore position data, to associate and map the boundary fracture parameter information and the wellbore position data to generate fracture wellbore associated data, and to measure bottom hole pressure of the standard fracturing horizontal well data based on the fracture wellbore associated data to obtain bottom hole pressure data.

[0148] An oil and gas non-steady state productivity index measurement module is configured to analyze mutual coupling of the boundary fracture characteristic information based on the bottom hole pressure data to generate boundary fracture relationship data, to identify oil and gas flow state of the oil and gas characteristic data based on the boundary fracture relationship data to obtain oil and gas flow state data, to measure oil and gas non-steady state productivity index of the standard fracturing horizontal well data based on the oil and gas flow state data to generate oil and gas non-steady state productivity index, to construct a non-steady state productivity prediction model using the oil and gas non-steady state productivity index to obtain the non-steady state productivity prediction model, and to predict non-steady state productivity of the standard fracturing horizontal well data based on the non-steady state productivity prediction model to generate a non-steady state productivity prediction report.

[0149] The application can ensure that original and direct information about the fractured horizontal well is obtained by using the sensor to collect data of the closed boundary fractured horizontal well through the data collection module, and can provide basic information for subsequent analysis; the data of the fractured horizontal well is preprocessed, which helps to eliminate noise and abnormal values in the data, can ensure the consistency and accuracy of the data, and provides key data input for subsequent determination of the fractured porosity and simulation of the oil and gas characteristics. The closed boundary fracturing characteristics of the standard fractured horizontal well data are identified through the closed boundary fracturing oil and gas simulation module, which can evaluate the key of the fracture network and the reservoir connectivity; the fracture characteristics of the closed boundary fracturing data are extracted, which can clearly determine the boundary fracture characteristic information; the fractured porosity of the standard fractured horizontal well data is determined according to the boundary fracture characteristic information, which can accurately reflect the pore structure of the fractured region, and is crucial for evaluating the reservoir characteristics of the oil and gas reservoir; the oil and gas characteristics of the standard fractured horizontal well data are simulated based on the boundary fracture characteristic information and the fractured porosity data, which can simulate the distribution and flow of oil and gas in the fractured horizontal well. The fracture parameters of the boundary fracture characteristic information are extracted through the oil and gas characteristic data through the fractured horizontal well bottom pressure measurement module, which can describe the geometric characteristics and distribution characteristics of the fracture in detail; the wellbore position of the standard fractured horizontal well data is identified according to the boundary fracture parameter information, which can determine the accurate position of the wellbore in the oil and gas reservoir; the boundary fracture parameter information is associated and mapped with the wellbore position data, which can associate the fracture parameters with the wellbore position, and provides data support for the bottom hole pressure measurement; the bottom hole pressure of the standard fractured horizontal well data is measured based on the fracture wellbore associated data, which can clearly determine the bottom hole pressure data. According to the bottom hole pressure data, the boundary fracture characteristic information is analyzed for the mutual coupling of the boundary fractures through the oil and gas unsteady-state productivity index measurement module, which can reveal the interaction and influence between the fractures; the oil and gas flow state of the oil and gas characteristic data is identified according to the boundary fracture relationship data, which can identify the flow state of the oil and gas in the fracture network; the oil and gas unsteady-state productivity index of the standard fractured horizontal well data is calculated through the oil and gas flow state data, which can provide dynamic change information of the oil and gas production; the unsteady-state productivity prediction model is constructed by using the oil and gas unsteady-state productivity index, and the unsteady-state productivity prediction model is obtained; the unsteady-state productivity of the standard fractured horizontal well data is predicted based on the unsteady-state productivity prediction model, and the unsteady-state productivity prediction report is generated. Therefore, through the data processing technology, the pattern recognition technology and the deep learning technology, the application realizes the fracture mutual relationship identification of the bottom hole pressure and the boundary fracture characteristics; and realizes the oil and gas flow state detection of the oil and gas characteristics according to the boundary fracture relationship, thereby improving the accuracy of the unsteady-state productivity prediction.

[0150] Therefore, the embodiments should be regarded, at any point, as being exemplary and not limiting, the scope of the application being defined by the appended claims and not by the above description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

[0151] The foregoing is considered as illustrative only of the principles of the application. Numerous modifications and changes will readily occur to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Therefore, the application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for predicting unsteady-state deliverability of a closed- boundary fractured horizontal well, characterized in that, The method comprises the following steps: Step S1: data acquisition is performed on the closed boundary fracturing horizontal well to generate fracturing horizontal well data; the fracturing horizontal well data is preprocessed to obtain standard fracturing horizontal well data; Step S2: closed boundary fracturing feature recognition is performed on the standard fracturing horizontal well data to obtain closed boundary fracturing data; fracture feature extraction is performed on the closed boundary fracturing data to generate boundary fracture feature information; fracturing porosity determination is performed on the standard fracturing horizontal well data according to the boundary fracture feature information to obtain fracturing porosity data; oil and gas feature simulation is performed on the standard fracturing horizontal well data based on the boundary fracture feature information and the fracturing porosity data to generate oil and gas feature data; Step S3: fracture parameter extraction is performed on the boundary fracture feature information through the oil and gas feature data to obtain boundary fracture parameter information; wellbore position recognition is performed on the standard fracturing horizontal well data according to the boundary fracture parameter information to obtain wellbore position data; fracture wellbore correlation data is generated by correlating and mapping the boundary fracture parameter information and the wellbore position data; wellbore pressure determination is performed on the standard fracturing horizontal well data based on the fracture wellbore correlation data to obtain wellbore pressure data; Step S4: boundary fracture mutual coupling analysis is performed on the boundary fracture feature information according to the wellbore pressure data to generate boundary fracture relationship data; oil and gas flow state recognition is performed on the oil and gas feature data according to the boundary fracture relationship data to obtain oil and gas flow state data; oil and gas non-steady-state productivity index measurement is performed on the standard fracturing horizontal well data through the oil and gas flow state data to generate oil and gas non-steady-state productivity index; a non-steady-state productivity prediction model is constructed by using the oil and gas non-steady-state productivity index to obtain a non-steady-state productivity prediction model; non-steady-state productivity prediction is performed on the standard fracturing horizontal well data based on the non-steady-state productivity prediction model to generate a non-steady-state productivity prediction report.

2. The method of predicting unsteady deliverability of a closed- boundary fractured horizontal well according to claim 1, wherein, Step S2 comprises the following steps: Step S21: non-linear feature decomposition is performed on the standard fracturing horizontal well data to generate non-linear fracturing horizontal well features; fracturing feature enhancement is performed on the non-linear fracturing horizontal well features to obtain fracturing feature enhancement data; closed boundary fracturing feature extraction is performed on the fracturing feature enhancement data to obtain closed boundary fracturing data; Step S22: fracturing morphology feature extraction is performed on the closed boundary fracturing data to generate fracturing morphology feature data; fracturing morphology geometric feature is obtained by performing geometric feature quantization on the standard fracturing horizontal well data; boundary fracture feature information is generated by performing fracture feature recognition on the fracturing morphology geometric feature; Step S23: distribution range determination is performed on the boundary fracture feature information to obtain a boundary fracture distribution range; fracture spatial distribution feature is generated by performing fracture spatial distribution feature recognition on the boundary fracture distribution range; fracture spatial distribution morphology feature is obtained by performing fracture spatial distribution morphology feature on the fracturing morphology geometric feature according to the fracture spatial distribution feature; Step S24: crack density calculation is performed on the crack spatial distribution pattern feature to generate crack spatial density data; based on the crack spatial density data, fracture pore feature is performed on the boundary crack feature information to obtain the fracture pore feature; and through the fracture pore feature, fracture porosity measurement is performed on the standard fracturing horizontal well data to obtain the fracture porosity data; Step S25: the fracture porosity data is divided by porosity gradient to generate porosity gradient data; based on the porosity gradient data and the boundary crack feature information, oil and gas feature simulation is performed on the standard fracturing horizontal well data to generate oil and gas feature data.

3. The method of predicting unsteady deliverability of a closed- boundary fractured horizontal well according to claim 1, wherein, Step S3 includes the following steps: Step S31: crack geometry feature recognition is performed on the boundary crack feature information to generate crack geometry feature data; and according to the oil and gas feature data, crack geometry parameter extraction is performed on the crack geometry feature data to obtain the crack geometry parameter; Step S32: crack connectivity detection is performed on the boundary crack feature information through the crack geometry parameter to generate crack connectivity data; based on the crack connectivity data, crack permeability evaluation is performed on the boundary crack feature information to obtain crack permeability data; and according to the crack permeability data, crack fluid flow simulation is performed on the oil and gas feature data to obtain crack fluid flow simulation data; Step S33: the boundary crack parameter information is obtained by integrating the crack connectivity data, the crack permeability data and the crack fluid flow simulation data; wellbore feature recognition is performed on the standard fracturing horizontal well data to generate wellbore feature data; and according to the boundary crack parameter information, wellbore position detection is performed on the wellbore feature data to obtain wellbore position data; Step S34: the boundary crack parameter information is associated with the wellbore position data to generate crack wellbore association data; and based on the crack wellbore association data, bottom hole pressure measurement is performed on the standard fracturing horizontal well data to obtain bottom hole pressure data.

4. The method of predicting unsteady deliverability of a closed- boundary fractured horizontal well according to claim 3, wherein, Step S34 includes the following steps: Step S341: connectivity quantification is performed on the crack connectivity data to obtain crack connectivity quantitative data; and crack connectivity association is performed between the crack connectivity quantitative data and the wellbore position data to generate crack wellbore connectivity association data; Step S342: permeability parameterization is performed on the crack permeability data to obtain crack permeability parameterized data; and crack permeability association is performed between the crack permeability parameterized data and the wellbore position data to generate crack wellbore permeability association data; Step S343: flow characteristic extraction is performed on the crack fluid flow simulation data to obtain crack fluid flow characteristic data; and fluid flow characteristic association is performed between the crack fluid flow characteristic data and the wellbore position data to generate crack wellbore flow association data; Step S344: the crack wellbore association data is obtained by integrating the crack wellbore connectivity association data, the crack wellbore permeability association data and the crack wellbore flow association data; Step S345: based on the crack wellbore association data, bottom hole pressure measurement is performed on the standard fracturing horizontal well data to obtain bottom hole pressure data.

5. The method of predicting unsteady deliverability of a closed- boundary fractured horizontal well according to claim 4, wherein, Step S345 includes the following steps: Step S3451: based on the fracture wellbore correlation data, the bottom hole stress feature of the standard fracturing horizontal well data is extracted, and the bottom hole stress feature is generated; according to the bottom hole stress feature, the bottom hole pressure distribution simulation of the standard fracturing horizontal well data is simulated, and the bottom hole pressure distribution simulation data is obtained; Step S3452: the bottom hole pressure response data is generated by identifying the bottom hole pressure response of the bottom hole pressure distribution simulation data; the bottom hole pressure numerical information is obtained by calculating the pressure value of the bottom hole pressure distribution simulation data; the bottom hole pressure gradient data is generated by dividing the pressure gradient of the bottom hole pressure numerical information; Step S3453: the bottom hole pressure change data is obtained by monitoring the real-time bottom hole pressure of the bottom hole pressure distribution simulation data through the bottom hole pressure gradient data; the bottom hole pressure state information is generated by identifying the bottom hole pressure state of the bottom hole pressure change data; Step S3454: the bottom hole pressure anomaly data is obtained by detecting the bottom hole pressure anomaly of the bottom hole pressure state information; the bottom hole pressure stability data is obtained by evaluating the pressure stability of the bottom hole pressure distribution simulation data according to the bottom hole pressure anomaly data; the bottom hole pressure data is obtained by measuring the bottom hole pressure of the standard fracturing horizontal well data based on the bottom hole pressure stability data.

6. The method of predicting unsteady deliverability of a closed- boundary fractured horizontal well according to claim 1, wherein, Step S4 includes the following steps: Step S41: according to the bottom hole pressure data, the boundary fracture feature information is analyzed, and the boundary fracture relationship data is generated; Step S42: according to the boundary fracture relationship data, the oil and gas flow state data is obtained by identifying the oil and gas flow state of the oil and gas feature data; Step S43: the oil and gas non-steady state productivity index is generated by calculating the oil and gas non-steady state productivity index of the standard fracturing horizontal well data through the oil and gas flow state data; Step S44: the non-steady state productivity prediction model is constructed according to the oil and gas non-steady state productivity index, and the non-steady state productivity prediction training model is obtained by training the non-steady state productivity prediction model using the oil and gas non-steady state productivity index; Step S45: the model evaluation data is obtained by evaluating the model cross-validation of the non-steady state productivity prediction training model; the non-steady state productivity prediction model is obtained by adjusting the model parameters of the non-steady state productivity prediction training model through the model evaluation data; Step S46: based on the non-steady state productivity prediction model, the non-steady state productivity prediction of the standard fracturing horizontal well data is generated, and the non-steady state productivity prediction report is generated.

7. The method of predicting unsteady deliverability of a closed- boundary fractured horizontal well according to claim 6, wherein, Step S41 includes the following steps: Step S411: the bottom hole pressure distribution feature is obtained by extracting the pressure distribution feature of the bottom hole pressure data; the bottom hole pressure peak area is generated by identifying the pressure peak area of the bottom hole pressure distribution feature; Step S412: the boundary fracture concentrated area is obtained by identifying the fracture concentrated area of the boundary fracture feature information according to the fracture space density data; the fracture pressure mapping data is generated by mapping and matching the boundary fracture concentrated area and the bottom hole pressure peak area; Step S413: Calculate the fracture pressure influence degree of the bottom hole pressure data by the fracture pressure mapping data, to obtain the fracture pressure influence data; determine the fracture sensitivity of the boundary fracture feature information according to the fracture pressure influence data, to generate the fracture sensitivity data; Step S414: Extract the boundary fracture constraint feature of the fracture sensitivity data by using the fracture connectivity data, to obtain the boundary fracture constraint action data; extract the boundary fracture permeability synergy feature of the boundary fracture constraint action data by using the fracture permeability data, to generate the boundary fracture synergy action data; identify the mutual coupling of the boundary fracture feature information according to the boundary fracture constraint action data and the boundary fracture synergy action data, to obtain the boundary fracture relationship data.

8. The method of predicting unsteady deliverability of a closed- boundary fractured horizontal well according to claim 6, wherein, Step S42 includes the following steps: Step S421: Correspond the fracture-oil and gas flow action of the oil and gas feature data according to the boundary fracture relationship data, to obtain the fracture-oil and gas flow action data; Step S422: Mark the oil and gas flow feature of the oil and gas feature data, to generate the oil and gas flow marking data; divide the flow unit of the oil and gas feature data by using the oil and gas flow marking data, to obtain the flow unit division data; extract the flow characteristics of the flow unit division data, to generate the oil and gas flow characteristic data; Step S423: Construct the fracture oil and gas flow network of the oil and gas flow characteristic data based on the fracture-oil and gas flow action data, to obtain the fracture oil and gas flow network; identify the oil and gas flow path of the fracture oil and gas flow network, to generate the oil and gas flow path data; measure the oil and gas flow rate of the fracture oil and gas flow network according to the oil and gas flow path data, to obtain the oil and gas flow rate data; Step S424: Detect the flow state anomaly of the fracture oil and gas flow network by using the oil and gas flow rate data, to generate the oil and gas flow state anomaly data; evaluate the flow state stability of the fracture oil and gas flow network, to obtain the flow state stability data; Step S425: Integrate the oil and gas flow rate data, the oil and gas flow state anomaly data and the flow state stability data, to obtain the oil and gas flow state data.

9. The method of predicting unsteady deliverability of a closed- boundary fractured horizontal well according to claim 6, wherein, Step S43 includes the following steps: Step S431: Calculate the unit oil and gas volume data by using the oil and gas flow rate data; generate the instantaneous oil and gas production data by using the unit oil and gas volume data; Step S432: Identify the flow anomaly state type of the oil and gas flow state anomaly data, to obtain the flow anomaly state type; record the abnormal instantaneous production of the instantaneous oil and gas production data according to the flow anomaly state type, to generate the abnormal state instantaneous production data; accumulate the oil and gas production of the abnormal state instantaneous production data, to obtain the abnormal state cumulative production data; Step S433: Record the stable state instantaneous production of the instantaneous oil and gas production data based on the flow state stability data, to generate the stable state instantaneous production data; accumulate the oil and gas production of the stable state instantaneous production data, to obtain the stable state cumulative production data; Step S434: Subtract the stable state cumulative production data from the abnormal state cumulative production data, to obtain the oil and gas non-steady state productivity index.

10. An apparatus for predicting unsteady-state deliverability of a closed- boundary fractured horizontal well, the apparatus comprising: The method comprises a prediction system for performing the method for predicting the unsteady productivity of a closed boundary fractured horizontal well according to claim 1, the prediction system comprising: a data acquisition module for acquiring data of the closed boundary fractured horizontal well to generate fractured horizontal well data, and preprocessing the fractured horizontal well data to obtain standard fractured horizontal well data; a boundary fractured oil and gas simulation module for identifying the closed boundary fracture characteristics of the standard fractured horizontal well data to obtain closed boundary fractured data, extracting the fracture characteristics of the closed boundary fractured data to generate boundary fracture characteristic information, determining the fracture porosity of the standard fractured horizontal well data according to the boundary fracture characteristic information to obtain fracture porosity data, and simulating the oil and gas characteristics of the standard fractured horizontal well data based on the boundary fracture characteristic information and the fracture porosity data to generate oil and gas characteristic data; a fractured horizontal well bottom pressure determination module for extracting the fracture parameters of the boundary fracture characteristic information through the oil and gas characteristic data to obtain boundary fracture parameter information, identifying the wellbore position of the standard fractured horizontal well data according to the boundary fracture parameter information to obtain wellbore position data, correlating and mapping the boundary fracture parameter information and the wellbore position data to generate fracture wellbore correlation data, and determining the bottom hole pressure of the standard fractured horizontal well data based on the fracture wellbore correlation data to obtain bottom hole pressure data; an oil and gas unsteady productivity index calculation module for analyzing the mutual coupling of the boundary fractures of the boundary fracture characteristic information according to the bottom hole pressure data to generate boundary fracture relationship data, identifying the oil and gas flow state of the oil and gas characteristic data according to the boundary fracture relationship data to obtain oil and gas flow state data, calculating the oil and gas unsteady productivity index of the standard fractured horizontal well data through the oil and gas flow state data to generate the oil and gas unsteady productivity index, constructing an unsteady productivity prediction model using the oil and gas unsteady productivity index to obtain the unsteady productivity prediction model, and predicting the unsteady productivity of the standard fractured horizontal well data based on the unsteady productivity prediction model to generate an unsteady productivity prediction report.

Images